Several weeks ago, the BBC and The Washington Post (among other outlets) reported on a 2018 study that seemed to imply that mobile phone use was causing “horns” to grow out of millennials’ heads. The problem? Well, the study was rife with issues, summarized well by PBS NewsHour science producer Nsikan Akpan in an article and on Twitter. Reporters, scientists and science readers alike should be worried about falling into this trap again. Here’s a quick checklist we can all use to improve our scientific literacy:
•Read the whole study: The abstract is often the easiest part of a paper to understand, but it’s also where the authors put the most spin on their findings. The methods and results sections will give a more complete picture of the study.
•Think critically about the sample: Sample size is typically disclosed in the introduction or methods section and is presented as a number (n = sample size). Do a quick gut check to make sure the number doesn’t seem too small, then look at what types of people or other sampling units were studied. In the bone spurs study, for example, all participants had gone to a chiropractor’s office for help, meaning that the sample wasn’t representative of a general population. It didn't include people who don't need a chiropractor.
•Learn some basic stats: P-hacking is a real concern. It’s helpful to know a bit about the statistical tests behind a “significant” result.
•Do a quick Google search on the authors: A Nobel laureate? A disgraced researcher? A chiropractor with a business selling posture pillows? You won’t know until you Google.
•Get an expert opinion: Be aware of when you’re out of your depth, and ask for help! The authors of similar studies, or members of 500 Women Scientists are great places to start.